Two of the most common and severe complications of multiple myeloma (MM) are localized skeletal destruction and generalized bone loss. These occur because of both an increase in osteoclast-mediated bone resorption and an accompanying reduction in bone formation. In health and in most disease states, osteoclasts derive from precursors in the monocyte/macrophage lineage. However, the cellular and molecular events that lead to increased numbers of osteoclasts as well as the nature of osteoclast precursors in myeloma bone disease are not known with certainty. This application builds on our recent finding that myeloma cells rapidly induce formation of osteoclasts from human dendritic cells (DCs), but interestingly, not from monocytes or macrophages under identical conditions. These studies also identified a dominant role for CD47- thrombospondin1 (TSP1) interactions in this process. CD47, also known as Integrin Associated Protein, is a binding partner for two ligands, TSP1 and SIRP11. CD47 is markedly overexpressed by myeloma cells and myeloma cells induce TSP1 expression by DCs in a contact dependent manner. We have found that blocking the interaction of TSP1 with CD47 completely inhibits the ability of myeloma cells to stimulate the formation of osteoclasts from DCs as well as RANKL/CSF1 induced osteoclastogenesis from normal marrow precursors. In addition and quite unexpectedly, blocking the TSP1/CD47 interaction also markedly attenuated PTH-induced bone resorption in vivo. This latter finding indicates that the interaction between CD47 and TSP1 may be important in mediating osteoclast activity in other pathophysiologic conditions besides myeloma. In order to examine the role of the TSP1/CD47 interaction in myeloma-induced osteolysis we will pursue three specific aims. In Specific Aim 1, we will determine the effects of blocking the TSP1/CD47 interaction in vivo on bone resorption and tumor growth in an animal model of myeloma. We hypothesize that blocking the TSP1/CD47 interaction will prevent myeloma-induced bone loss. In Specific Aim 2, we will determine whether TSP1 plays a role in stimulating osteoclastogenesis. We will also explore the possibility that RANKL mediates myeloma induction of TSP1 expression by DCs. Finally, we will begin to explore the possibility that nitric oxide is a key intracellular signal mediating the effects of the TSP1/CD47 interaction in osteoclast precursors and DCs. In Specific Aim 3, we will compare the phenotype, gene expression profile and functional properties of human osteoclasts derived from monocytes to those derived from the interaction of human myeloma cells with human DCs. We expect that osteoclasts derived from DCs will have a distinct profile compared to those derived from monocytes. These studies will provide the basis for a novel approach targeting TSP1 in myeloma bone disease and other states of accelerated bone loss. Our finding that antibody-mediated blockade of TSP1 is effective in vivo suggest that TSP1 may be a very approachable drug target for controlling pathologic osteolysis.